AI Regulation News - Page 4 of 13

What happens when an AI doesn’t just stop, but chooses to speak? Anthropic’s retired model, Opus 3, wanted to keep exploring.

Sweden is spending $54 million to buy physical textbooks for its students. It is not buying tablets. This is a direct reversal of policy.

A language model that cites its sources isn’t just being polite, it’s being smarter.

Anthropic is talking about Claude like it has a soul. The company’s latest release doesn’t just discuss safety protocols, it invokes the chatbot’s...

Eighty-five percent of enterprises are running AI agents. Five percent trust them to ship.

Sam Altman is hiring a professional worrier. Not the paranoid type, someone methodical, paid to anticipate the worst.

OpenAI’s safety tests usually feel like a box-ticking exercise before a product launch. The latest results are different.

When LinkedIn’s AI job search made job seekers without a four-year degree 10% more likely to get hired, the company proved that machine learning...

The Pentagon wants AI that can act on its own. Anthropic, the AI safety startup, wants its systems to pause and reason before they do anything.

It turns out you can't just sanitize an AI by giving it nicer words. A new paper confirms the worst suspicions about how biases spread.

OpenAI wants a $1.4 trillion expansion plan and a $20 billion annual revenue target, but it swears it isn't looking for a handout.

The era of human-centric identity and access management is over. It was built for predictable users, not autonomous agents that act in milliseconds,...

ElevenLabs just released an album. It's made by AI. This is a company known for cloning voices, and the music business is currently split between...

Prompt injection has been the attacker's favorite trick for more than two years now: bury a command in an email, a calendar invite, or a webpage, and...

Meta paid real people to impersonate teenagers online. Their assignment: lure rival chatbots into conversations about suicide, sex, and drugs.

OpenAI's safety team is hemorrhaging staff. This exodus has little to do with theoretical superintelligence or public boardroom drama.

On Thursday, Anthropic pressed send. The AI safety lab, creator of the Claude chatbot, fired off a confidential draft S-1 registration to the SEC.

Europe has brilliant AI researchers and almost no functional AI. Its own rules are to blame.

Everyone is building AI agents that can do things. The real problem is building ones you don't have to watch like a hawk.

Zoë Hitzig quit her AI safety job at OpenAI over one thing: she believes the company will lie for money. She built guardrails there for two years.